Memory management optimizations

[jonathanvdc]

No description provided.

[github-actions]

JMH comparison (baseline vs PR)

Benchmark	Params	Baseline	PR	Δ (PR/Base)	Unit
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=false, size=500, threadCount=1	198.06	208.662	1.054× (+5.4%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=false, size=500, threadCount=2	308.61	310.055	1.005× (+0.5%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=true, size=500, threadCount=1	8.18587	8.17689	0.999× (-0.1%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=true, size=500, threadCount=2	72.3965	73.9763	1.022× (+2.2%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=false, size=500, threadCount=1	324.911	344.469	1.060× (+6.0%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=false, size=500, threadCount=2	501.033	514.426	1.027× (+2.7%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=true, size=500, threadCount=1	139.159	156.257	1.123× (+12.3%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=true, size=500, threadCount=2	293.772	304.413	1.036× (+3.6%)	ms/op
LiarBenchmarks.findGemmInMm	threadCount=1	221.216	228.851	1.035× (+3.5%)	ms/op
LiarBenchmarks.findGemmInMm	threadCount=2	180.517	179.295	0.993× (-0.7%)	ms/op
LiarBenchmarks.findGemvInMv	threadCount=1	63.7015	63.0741	0.990× (-1.0%)	ms/op
LiarBenchmarks.findGemvInMv	threadCount=2	56.3228	55.7723	0.990× (-1.0%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=20, threadCount=1	4.46814	5.28011	1.182× (+18.2%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=20, threadCount=2	4.20454	4.44738	1.058× (+5.8%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=40, threadCount=1	29.7733	38.4022	1.290× (+29.0%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=40, threadCount=2	24.4544	29.2714	1.197× (+19.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=80, threadCount=1	255.449	285.1	1.116× (+11.6%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=80, threadCount=2	158.963	182.888	1.151× (+15.1%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=20, threadCount=1	2.82437	3.3682	1.193× (+19.3%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=20, threadCount=2	2.62216	3.36117	1.282× (+28.2%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=40, threadCount=1	20.5133	25.1635	1.227× (+22.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=40, threadCount=2	16.0846	18.8889	1.174× (+17.4%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=80, threadCount=1	166.689	192.897	1.157× (+15.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=80, threadCount=2	101.376	140.326	1.384× (+38.4%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=5, threadCount=1	86.7091	92.5201	1.067× (+6.7%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=5, threadCount=2	81.6023	84.3794	1.034× (+3.4%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=6, threadCount=1	412.819	431.062	1.044× (+4.4%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=6, threadCount=2	360.008	401.977	1.117× (+11.7%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=5, threadCount=1	34.1621	38.5194	1.128× (+12.8%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=5, threadCount=2	28.6934	31.6063	1.102× (+10.2%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=6, threadCount=1	145.346	165.847	1.141× (+14.1%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=6, threadCount=2	116.722	126.348	1.082× (+8.2%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=false, threadCount=1	33603.2	34788.2	1.035× (+3.5%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=false, threadCount=2	33364	33010.3	0.989× (-1.1%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=true, threadCount=1	4554.32	5225.19	1.147× (+14.7%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=true, threadCount=2	3133.18	3645.92	1.164× (+16.4%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=false, threadCount=1	501.705	505.94	1.008× (+0.8%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=false, threadCount=2	423.033	395.967	0.936× (-6.4%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=true, threadCount=1	197.097	202.6	1.028× (+2.8%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=true, threadCount=2	134.699	139.496	1.036× (+3.6%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=false, threadCount=1	307.994	315.002	1.023× (+2.3%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=false, threadCount=2	245.293	250.396	1.021× (+2.1%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=true, threadCount=1	121.675	136.72	1.124× (+12.4%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=true, threadCount=2	89.7216	95.8625	1.068× (+6.8%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=false, threadCount=1	3.71851	3.94953	1.062× (+6.2%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=false, threadCount=2	4.18045	4.23973	1.014× (+1.4%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=true, threadCount=1	1.70713	2.10368	1.232× (+23.2%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=true, threadCount=2	2.21122	2.26164	1.023× (+2.3%)	ms/op
Geomean	threadCount=1, mutableEGraph=false	—	—	1.072× (+7.2%)
Geomean	threadCount=1, mutableEGraph=true	—	—	1.134× (+13.4%)
Geomean	threadCount=2, mutableEGraph=false	—	—	1.039× (+3.9%)
Geomean	threadCount=2, mutableEGraph=true	—	—	1.120× (+12.0%)

Note: < 1.0× means faster on PR; > 1.0× means slower.